Many different types of memory exist to store data for computers and other digital devices. Flash memory is one type of memory that has become popular because it combines the high density and low cost of EPROM with the electrical erasability of EEPROM. Flash memory can be rewritten and can hold its contents without power, and thus is nonvolatile. It is used in many portable electronic products, such as cell phones, portable computers, voice recorders, etc., as well as in many larger electronic systems, such as cars, planes, industrial control systems, etc.
FIG. 1 shows one example of a dual-bit flash memory cell 100, which is a memory cell is essentially split into two identical (mirrored) parts, each of which is formulated for storing one of two independent bits. Thus, one bit of data can be independently stored in one of two storage regions 102, 104. The dual bit memory cell, like a traditional single bit flash memory cell, has a wordline (WL) 106 associated with a source 108 and a drain 110. Like a traditional single bit flash memory cell, a channel region 112 may exist under the wordline between the source and drain.
A stacked gate structure 114, which may include a multi-level dielectric layer 116 and the wordline 106, is formed over the semiconductor substrate or p-well 118. The multi-level dielectric layer 116 can generally be composed of three separate layers: a first insulating layer 120, a charge-trapping layer 122, and a second insulating layer 124. The wordline 106 is formed over the multi-layer dielectric layer 116.
In FIGS. 2A-2D, a prior art method 200 for performing a memory operation on the dual-bit memory cell 100 is described with reference to a flow diagram (FIG. 2A) and schematic representations (FIGS. 2B-2D). The illustrated flow diagram (FIG. 2A) shows general functional steps, while the schematic representations (FIG. 2B-2D) show various bias conditions applied on a dual-bit cell and movement of charged carriers (i.e., holes p+, and/or electrons n−) at various blocks of the method 200.
As shown in FIG. 2A, the method 200 proceeds differently depending on whether a user is programming the memory cell or erasing the memory cell. If the operation is a program operation (“PROGRAM” at 202), the method 200 proceeds to block 204 wherein one bit of the memory cell is programmed. Conversely, if the operation is an erase operation (“ERASE” at 202), the method proceeds to block 206 wherein one bit of the memory cell is erased.
After the right bit of the cell is programmed (FIG. 2B, 204A) and/or the left of the cell is programmed (FIG. 2C, 204B), a user may want to erase either bit of data in the cell. Referring now to FIG. 2D, one can see an erase operation 206 in accordance with the prior art, wherein the right bit of the cell is erased. A relatively high voltage (6V) is applied to the drain, and the wordline WL is held at a negative voltage (−6V), while the source is allowed to float. Under these conditions, a strong electric field is developed across the multi-layer dielectric 116 between the wordline and the drain. The negatively charged electrons that are trapped in the charge trapping layer 122 are attracted to the positive voltage on the drain, while positively charged holes flow from the substrate or p-well 118 into the charge trapping layer 122, both by way of Fowler-Nordheim tunneling through the tunnel oxide 120. As the electrons are removed from the charge trapping layer and holes flow into the charge trapping layer, the cell is erased.
One shortcoming of the prior art method 200 is that hot-carriers (e.g., accelerated carriers) are difficult to precisely control. Thus, for example, while the cell is erased, an excess number of holes may be injected into the charge trapping layer 122 or an excess number of electrons may be removed from the charge trapping layer. In either case, the voltage threshold of the cell may differ from an expected threshold value, causing the data stored in the cell to be less reliable. Therefore, during a subsequent read operation, the cell may provide a different current from the drain to source then expected. Because the current level may be indicative as to the data value stored in the cell, the cell will return unreliable data.
Therefore, a need has arisen to provide systems and methods relating to memory devices that cure some deficiencies of the prior art.